Generation of Auxotrophic Mutants of Enterococcus faecalis

JOURNAL OF BACTERIOLOGY, Dec. 1995, p. 6866–6873 0021-9193/95/$04.0010 Copyright q 1995, American Society for Microbiology

Vol. 177, No. 23

Generation of Auxotrophic Mutants of Enterococcus faecalis XIAOTAO LI,1 GEORGE M. WEINSTOCK,1,2

AND

BARBARA E. MURRAY1,3*

1

Department of Microbiology and Molecular Genetics, Department of Biochemistry and Molecular Biology,2 and Division of Infectious Diseases, Department of Internal Medicine,3 University of Texas Medical School, Houston, Texas 77030 Received 6 July 1995/Accepted 21 September 1995

A 22-kb segment of chromosomal DNA from Enterococcus faecalis OG1RF containing the pyrimidine biosynthesis genes pyrC and pyrD was previously detected as complementing Escherichia coli pyrC and pyrD mutations. In the present study, it was found that the E. faecalis pyrimidine biosynthetic genes in this clone (designated pKV48) are part of a larger cluster resembling that seen in Bacillus spp. Transposon insertions were isolated at a number of sites throughout the cluster and resulted in loss of the ability to complement E. coli auxotrophs. The DNA sequences of the entire pyrD gene of E. faecalis and selected parts of the rest of the cluster were determined, and computer analyses found these to be similar to genes from Bacillus subtilis and Bacillus caldolyticus pyrimidine biosynthesis operons. Five of the transposon insertions were introduced back into the E. faecalis chromosome, and all except insertions in pyrD resulted in pyrimidine auxotrophy. The prototrophy of pyrD knockouts was observed for two different insertions and suggests that E. faecalis is similar to Lactococcus lactis, which has been shown to possess two pyrD genes. A similar analysis was performed with the purL gene from E. faecalis, contained in another cosmid clone, and purine auxotrophs were isolated. In addition, a pool of random transposon insertions in pKV48, isolated in E. coli, was introduced into the E. faecalis chromosome en masse, and an auxotroph was obtained. These results demonstrate a new methodology for constructing defined knockout mutations in E. faecalis. Enterococci, which have been recognized as a cause of many infections, including infectious endocarditis (26), are the secondto third-most-common pathogens found in hospital-acquired infections. Because of an alarming increase in resistance to different antibiotics, therapy of enterococcal infections has become increasingly difficult. Thus, there is a need for more basic knowledge of these organisms in order to understand how to control or prevent enterococcal infections through developing either therapeutics or vaccines. In a previous report, we described the physical map of the genome of Enterococcus faecalis OG1RF as well as several cosmid clones that complemented Escherichia coli auxotrophs (27). In this study, the possibility of generating enterococcal pyr and pur auxotrophs by allelic replacement was tested. This methodology is important for developing genetic approaches for studying enterococcal virulence. We also describe the further characterization of a clone containing pyr biosynthesis genes and the determination of the arrangement of the pyr gene cluster by mapping, sequencing, and complementation.

medium for growth of E. coli. For testing of possible enterococcal auxotrophs, Davis minimal medium with supplements (DMMS) as described by Murray et al. (27) was used as the defined synthetic enterococcal broth medium. DMMS agar contains 1.5% Bacto Agar (Difco Laboratories, Detroit, Mich.). Uracil or adenosine was added at a concentration of 40 mg/ml for the growth of auxotrophs in defined media. The media for electroporation included brain heart infusion (BHI; Difco), BYGT (BHI, yeast extract, glucose, Tris) (8), and SR (tryptone, yeast extract, sucrose, glucose, gelatin, agar) (7). Concentrations of antibiotics used for selection were as follows: tetracycline at 12.5 mg/ml, chloramphenicol at 20 to 40 mg/ml, ampicillin at 150 mg/ml, nalidixic acid at 20 mg/ml, and kanamycin at 25 mg/ml for E. coli and rifampin at 100 mg/ml, fusidic acid at 25 mg/ml, and kanamycin at 2,000 mg/ml for OG1RF. Routine DNA techniques. Plasmid DNA was isolated by a slightly modified alkaline sodium dodecyl sulfate protocol (4): solution I contained no lysozyme, and solution III was made of 3 M potassium acetate and 5 M glacial acetic acid. Preparation of competent cells and transformation of plasmid DNA were performed as described previously (6) or by electroporation (5) using a Bio-Rad Gene Pulser (Bio-Rad, Hercules, Calif.). Transducing lysate preparation and transduction were performed by using standard procedures (34). Southern transfer and hybridizations were carried out as described previously (27). PCR was performed by using a PCR Optimizer Kit (Invitrogen, San Diego, Calif.) and a DNA Thermal Cycler (Perkin-Elmer Cetus, Norwalk, Conn.). Complementation. The cosmid pKV48 was transferred into E. coli pyrimidine auxotrophic mutants by conjugative mobilization, transformation, or transduction. For conjugative mobilization, E. coli MG1655 containing pKV48, one of the E. coli chloramphenicol-resistant pyrimidine auxotrophs (GE1721 or GE1728; Table 1), and the mobilizing strain GE3422, which contains pRK2013 (10), were each grown in LB broth to log phase (about 4 h); 200-ml aliquots of the cultures were combined and incubated at 378C for 2 h without shaking. The mating mixture was diluted and spread onto LB-tetracycline-chloramphenicol. For E. coli pyrAa, pyrAb, pyrB, pyrF, and pyrE auxotrophs (Table 1), pKV48 was introduced by transformation or transduction, selecting for tetracycline resistance. The E. coli pyr auxotrophs containing pKV48 were scored for the ability to grow on the M63 minimal agar. pKV48 derivatives containing transposon insertions were also transduced into the seven E. coli pyr auxotrophs, selecting for both kanamycin and tetracycline resistance. Complementation was tested by comparing colony formation on the following media: M63 agar (30 and 378C), M63kanamycin and M63-uracil. pKV53 and its derivatives were tested similarly in an E. coli purL mutant (GE1726), with or without adenosine, for complementation. Transposon mutagenesis. To identify the positions of the pyr genes, pKV48 was mutagenized by using the transposons TnphoA and mgd. For mutagenesis with TnphoA, a fresh lTnphoA stock was prepared in LE392. E. coli MG1655 containing pKV48 was grown for about 3 h, and lTnphoA was added at a multiplicity of infection of approximately 1. Following incubation at 308C for 15 min, the cells were diluted 1:10 into LB broth to allow outgrowth for 4 h at 378C with aeration. Aliquots of 200 ml were plated on LB-kanamycin with 40 mg of XP

MATERIALS AND METHODS Bacterial strains, plasmids, and phage. The bacterial strains used in this work are described in Table 1. The cosmid vector pLAFRx is a derivative of pLAFR which contains oriT of RK2 and a polylinker for cloning (12, 17). pKV48 and pKV53 are cosmids from the previously constructed enterococcal genomic libraries from E. faecalis OG1RF (Table 1) which complement E. coli pyrC and purL auxotrophs, respectively (27). The vector used for subcloning was pBluescript II SK1 phagemid (36). Transposon mutagenesis was performed by using TnphoA (22) from a lTnphoA derivative that also carried the cI857 and P(Am)80 mutations and mini-gd-200 (mgd) from strain CBK884, obtained from Michael G. Caparon (11). Lambda cI857 was used for transduction in E. coli. Media. E. coli transformants and transconjugants were grown on LB agar (25) with appropriate antibiotics. M63 salts (25) supplemented with 0.2% glucose, thiamine (100 mg/ml), MgSO4 (1 mM), and 1.5% agar was used as the minimal

* Corresponding author. Mailing address: 1.728 JFB, Infectious Diseases, University of Texas Medical School, 6431 Fannin St., Houston, TX 77030. Phone: (713) 745-0131. Fax: (713) 745-0130. 6866

E. FAECALIS AUXOTROPHS

VOL. 177, 1995

6867

TABLE 1. Bacterial strains and plasmids used Strain or plasmid

Strains E. coli MG1655 LE392 KK2186 GE3422 W1485 GE1728 GE1721 GE1726 RC50 Hfr3000YA289 X82 AT2538 JEF8 CBK884 LW49 E. faecalis OG1RF TX5063a TX5066a TX5077 TX5078 TX5079 TX5082 TX5081 Plasmids pBluescript pLAFRx pKV48 pKV53 pBEM201-pBEM213 pBEM214 pBEM215 pBEM216 pBEM217 pBEM218 pBEM219 pBEM220 pBEM221 pBEM222 pTX5062 a

Relevant characteristics

Reference(s) and/or source

F 2 l2 e14-(mcrA) hsdR514 supE44 supF58 lacY1 or D(lacIZY)6 galK2 galT22 metB1 trpR55 endA1 hsdR4 supE sbcI5 thi-1 strA LE392 containing pRK2013, Kanr F1 l2 MG1655 pyrC::Tn10dCamNS Camr MG1655 pyrD::Tn10dCamNS Camr MG1655 purL::Tn10dCamNS Camr carA50 thi-1 malA xyl-7 strA135 lr l2 tsx-273 (also referred to as pyrAa) pyrB289 l2 relA1 spoT1 thi-1 pyrF287 lacZ53(Am) l2 trpC60 hisG1 rpsL8 pyrE60 l2 thr-1 ara-14 leuB6 D(gpt-proA)62 lacY1 supE44 galK2 hisG4 fbD1 rpsL31 xyl-5 mtl-1 argE3 carB8 thr-31 crelA1 metB1 (also referred to as pyrAb) mgd mutagenesis donor strain containing pMGD5 and pXRD4043, Kanr Camr mgd mutagenesis recipient strain, Nalr

16 34 18 10 2 27 27 27 24; CGSCa CGSC CGSC CGSC CGSC 11 11

OG1 Rifr Fusr (spontaneous) OG1RFpurL::mgd5062, generated by allelic replacement with pTX5602; auxotroph; Kanr Rifr Fusr OG1RFpyrD::mgd220, generated by allelic replacement with pTX5601; no auxotrophic phenotype; Kanr Rifr Fusr OG1RFpyrE::mgd217 generated by allelic replacement with pBEM217; auxotroph; Kanr Rifr Fusr OG1RFpyrC::mgd219, generated by allelic replacement with pBEM219; auxotroph; Kanr Rifr Fusr OG1RFpyrR::mgd218, generated by allelic replacement with pBEM218; auxotroph; Kanr Rifr Fusr OG1RFpyrD::mgd221, generated by allelic replacement with pBEM221; no auxotrophic phenotype; Kanr Rifr Fusr OG1RFpyr::mgdC22, generated by allelic replacement with a pool of pKV48::mgd; auxotroph; Kanr Rifr Fusr

27 This study This study

2,958-bp phagemid derived from pUC19, Ampr 21.6-kb cosmid vector with mob site and oriT of RK2, Tetr pLAFRx containing a 22-kb fragment from OG1RF with pyr gene cluster, Tetr pLAFRx containing a 24-kb fragment from OG1RF; purL complementing clone; Tetr pKV48 with TnphoA insertion at positions 201–213, Tetr Kanr 3-kb EcoRI-BamHI fragment of pKV48 cloned into pBluescript, Ampr Deletion of pBEM207, Tetr Kanr 2.5-kb NotI-HindIII fragment of pBEM207 cloned into pBluescript, Ampr pKV48pyrE::mgd217 Tetr Kanr pKV48pyrR::mgd218 Tetr Kanr pKV48pyrC::mgd219 Tetr Kanr pKV48pyrD::mgd220 Tetr Kanr (also called pTX5061) pKV48pyrD::mgd221 Tetr Kanr pKV48pyrAa::mgd222 Tetr Kanr pKV53purL::mgd5062 Tetr Kanr

36 27 27, this study 27, this study This study This study This study This study This study This study This study This study This study This study This study

This This This This

study study study study

This study

CGSC, E. coli Genetic Stock Center, Yale University.

(5-bromo-4-chloro-3-indolyl phosphate) per ml and incubated for 2 to 3 days at 378C. Blue colonies were saved separately, and then all colonies were pooled. Cosmids contained in the pool were mobilized to the chloramphenicol-resistant auxotrophic E. coli recipients by triparental mating with W1485 as the helper strain. Strains containing cosmids with a TnphoA insertion(s) were selected on LB-tetracycline-chloramphenicol-kanamycin plates, and then the transposon insertion in the cosmid was confirmed by restriction endonuclease digestion. Thirteen cosmids (named pBEM201 to pBEM213) with TnphoA insertions at different positions in pKV48 were used for subsequent analysis. To make knockout mutations for construction of subsequent enterococcal mutants with insertions in pyr or pur genes, the transposon mgd was chosen to mutagenize pKV48 and pKV53. The kanamycin resistance determinant in mgd is expressed in both gram-positive and gram-negative hosts and can thus be used as a selective marker for introducing insertions into enterococci. The mutagenesis strain, CBK884(pMGD5, pXRD4043), which contains mgd on a conjugative donor plasmid and a source of gd transposase, was transformed with pKV48 or pKV53 with selection on LB-tetracycline-kanamycin. A single transformant colony was then grown in 3 ml of LB-chloramphenicol (40 mg/ml) with 0.5 mM IPTG (isopropylthiogalactoside) for 3 h to induce transposase production; the recipient strain, LW49, was grown in LB-nalidixic acid. Donor cells (0.5 ml) and 0.2 ml of recipient cells were mixed in an 18-mm-diameter test tube and incubated for 30 min in a 378C water bath without shaking. Then 5 ml of prewarmed

LB broth containing IPTG was added, and the mixture was incubated for 3 h without agitation. Exconjugants were selected on LB-tetracycline-kanamycinnalidixic acid. The pools of exconjugants containing pKV48 or pKV53 with mgd insertions were transduced into several E. coli pyr auxotrophs or the pur auxotroph and purified for single colonies. Purified single colonies were then transferred with toothpicks onto both M63 agar and LB-tetracycline-kanamycin. Strains with mutant cosmids that no longer complemented the relevant E. coli auxotroph were saved for subsequent sequencing and allelic replacement studies. Transposon insertions were also mapped according to restriction endonuclease digestion patterns. Restriction mapping and subcloning. pKV48 and selected derivatives with TnphoA insertions were analyzed by digestion with EcoRI, BamHI, ClaI, and HindIII (Promega, Madison, Wis.), using buffers supplied with the enzymes or KGB reaction buffer (34). Digested plasmid DNA was electrophoresed in 0.6% ultraPURE electrophoresis-grade agarose (GIBCO BRL, Life Technologies Inc., Gaithersburg, Md.) in 13 TBE (0.09 M Tris base, 0.09 M boric acid, 0.002 M EDTA). The 1-kb DNA ladder and HindIII-digested l DNA fragments (GIBCO BRL) were used as molecular weight markers. Determination of fragment sizes by measurements of photographed gels and comparisons of endonuclease restriction patterns were used to generate a restriction map of pKV48. To generate subclones of pKV48, both pBluescript and pKV48 were double digested with 10 U each of EcoRI and BamHI (Promega) for 2 to 4 h in KGB

6868

LI ET AL.

buffer. Digested DNA was extracted with 1 volume of phenol-chloroformisoamyl alcohol (Amresco, Solon, Ohio), and the supernatant was precipitated with 2 volumes of 95% ethanol. DNA fragments excised from agarose gels were purified by electroelution (38). Digested pBluescript and pKV48 were then mixed and ligated with 40 U of T4 DNA ligase (New England Biolabs, Beverly, Mass.), using the reaction buffer supplied. All subclones were transformed into E. coli KK2186. Insertions in pBluescript were detected on LB-ampicillin with 2 mg of X-Gal (5-bromo-4-chloro-3-indolyl-b-D-galactoside) per plate. White colonies were randomly selected for plasmid extraction and digestions. DNA sequencing and analyses. One-pass DNA sequencing was performed for the pyr and purL genes, with the exception of pyrD, whose DNA sequence was determined until all ambiguities were resolved. The nucleotide sequence of part of the pyr cluster was determined by the dideoxy-chain termination method (35), using automated sequencing (32). Some of the cluster and most of the pyrD gene were also manually sequenced with 35S-labeled nucleotides. The Taq DyeDeoxy Terminator Cycle Sequencing Kit (Applied Biosystems Inc. [ABI], Foster City, Calif.) was used for PCR amplification according to the protocol provided with the kit. Samples were analyzed using an ABI model 373A DNA sequencer. Data were processed with a sequencing analysis program from ABI. The primers used for sequencing included pBluescript T7 and T3 primers (Stratagene, Menasha, Wis.), a primer which is complementary to both inverted repeat sequences of TnphoA (GW26, 59-ACG GGA AAG GAT CCG TCC AGG-39), a primer complementary to one end of mgd (mgd-R, 59-GAT TTA GGA TAC ACG GAA TTT CG-39), a primer complementary to the phoA gene in TnphoA (59-CTG AGC AGC CGG GTT TTC CAG AAC AG-39), and primers specific to sequences generated in this study. Both the phoA primer and the mgd-R primer, which is complementary to the mgd right-arm sequence, were used for sequencing DNA flanking the transposon. Cosmids pBEM220 (also called pTX5061), pTX5062, pBEM201, pBEM206, pBEM208, pBEM209, pBEM217 to pBEM219, pBEM221, and pBEM222 were used for direct sequencing without subcloning, using the phoA primer or the mgd-R primer. The GW26 primer was used when only one of the TnphoA inverted repeats was present such as in pBEM215 and pBEM216. Double-stranded cosmid DNA was prepared for automated sequencing using a Magic Minipreps DNA Purification System Sample Kit (Promega) or by equilibrium centrifugation in CsCl-ethidium bromide gradients (34). DNA sequence analysis was performed with the Genetics Computer Group (University of Wisconsin, Madison) sequence analysis package, version 7.2. DNA and protein homology searches were performed with either the BLAST or the FASTA/TFASTA sequence comparison algorithms (30). Both searches were performed via the GenEMBL and SWISS-PROT databases. Allelic replacement. Cosmids with mgd insertions that had lost the ability to complement an auxotroph of E. coli (including pBEM220, pTX5062, pBEM217 to pBEM219, and pBEM221) were transformed into OG1RF by using a modification of the electroporation protocol described previously (9). For preparation of competent cells, OG1RF was grown overnight in BYGT with various glycine concentrations. A concentration of around 6% glycine led to optimal growth inhibition (70 to 90% reduction in optical density at A660). The overnight culture was diluted 10-fold into fresh medium with the same concentration of glycine and incubated for 1 h at 378C. Cells were chilled on ice, harvested by centrifugation at 5,000 3 g for 12 min, and washed twice with 1/3 of the original volume of chilled electroporation solution (0.625 M sucrose–1 mM MgCl2 adjusted to pH 4.0 with 1 N HCl). The washed cells were resuspended in 1/30 of the original volume of electroporation solution and then incubated on ice for 30 to 60 min (or saved at 2708C for later use). Cosmid DNA was prepared by equilibrium centrifugation in CsCl-ethidium bromide gradients. Competent cells (100 to 200 ml) were mixed with 2 to 5 mg of DNA (in ,20 ml of distilled H2O or low-salt buffer), added to a chilled 0.2-cm cuvette, and electroporated immediately with a BioRad Gene Pulser apparatus at a capacitance of 25 mF, resistance of 200 V, and peak voltage of 2.5 kV (field strength of 8,750 to 10,000 V/cm for E. faecalis). Cells were then incubated in 1 ml of BYGT with 0.25 M sucrose for 90 to 120 min at 378C and plated on SR agar plates containing kanamycin (2,000 mg/ml). After electroporation, colonies growing on the selective plates were further tested by colony hybridization and by Southern hybridization to show the physical structure expected from allelic replacement by homologous recombination. To determine if pLAFRx sequences were present, colony hybridization was carried out as described previously (3), using the whole pLAFRx as a probe. The presence of mgd was verified by using a 2-kb BamHI fragment from mgd as a probe. The probe used for Southern hybridization of TX5066a chromosomal DNA was a 1-kb PCR product containing the pyrD gene generated with primers just upstream and downstream of pyrD. The probe for Southern hybridization of TX5063a chromosomal DNA was a 4-kb EcoRI fragment excised from pKV48. To simplify the process of screening inactivated pyr genes in E. coli and to identify possible mutations in regions other than structural genes, we tested the possibility of generating enterococcal auxotrophs with allelic replacement by electroporating a pool of different pKV48::mgd insertions. Insertions of mgd into pKV48 were selected as described above, and several hundred of the mutants were pooled. Cosmid DNA was prepared from the pooled cells as described above and used for electroporation of OG1RF. Transformants were selected on BHI-kanamycin (2,000 mg/ml) and saved for further analysis, including growth in DMMS broth and agar. Southern blotting was performed to map the insertions in OG1RF, with pKV48 as the probe.

J. BACTERIOL. Growth curves. The phenotype of the OG1RF::mgd derivatives created by allelic replacement was first screened on DMMS agar as described above and then tested in DMMS broth. For growth in broth, OG1RF strains were grown overnight in BHI broth, harvested, washed twice with 0.9% NaCl, and then added to the DMMS broth at a final inoculum of 107 CFU/ml. The cell densities of OG1RF and the OG1RF transposon insertion mutants were determined by measuring the optical density at A660 and Klett units. Growth of cultures with and without adenosine or uracil was assessed by Klett units and colony counts at various times over 18 h of incubation (shaken at 300 rpm) at 378C. Nucleotide sequence accession numbers. The sequences reported have been submitted to GenBank and assigned accession numbers as follows: pyrR, U25091; pyrP, U25095; pyrB, U25092; pyrC, U25093; pyrAb, U25090; pyrD, U24692; pyrF, U25094; pyrE, U24682; and pyrAa, U36195.

RESULTS Complementation and transposon mutagenesis. Previous studies have shown that a pyrC mutant of E. coli was complemented by the cosmid clone pKV48 (27). The possibility that pKV48 could complement other mutations in the pyrimidine biosynthesis pathway was tested by introducing this cosmid into various E. coli pyr mutants and scoring for their abilities to grow on minimal media (Table 2). Prototrophy was restored for E. coli pyrAa, pyrAb, pyrB, pyrC, pyrD, pyrF, and pyrE auxotrophs, indicating that genes encoding analogous products were contained in the cosmid. The E. coli pyrAa and pyrAb (also referred to as carA and carB) mutants require both arginine and uracil. These auxotrophs were complemented by pKV48 on minimal media in the presence of arginine but not in its absence or when only uracil was present. Although the pyrimidine biosynthesis functions are spread throughout the E. coli chromosome, they are clearly clustered in E. faecalis. This result is similar to that found for Bacillus species (14, 33). Mapping pyr genes by transposon mutagenesis. Thirteen TnphoA insertions and six mgd insertions were isolated and mapped in pKV48 (Fig. 1 and Table 1). Cosmids with representative insertions were tested for complementation in seven E. coli pyr gene mutants (Table 2). Among the TnphoA insertion derivatives of pKV48, pBEM207, pBEM208, and pBEM 209 (with insertions at positions 207, 208, and 209, respectively) had lost the ability to complement the pyrC, pyrAa, and pyrAb mutations. pBEM205 and pBEM206 had lost the ability to complement the E. coli pyrAa and pyrAb auxotrophs. pBEM 202 and pBEM203 had lost the ability to complement the pyrD mutation, while pBEM204 complemented pyrD and therefore may lie outside this gene. pBEM201 failed to complement the pyrF mutant. pBEM210 complemented pyrC, suggesting that the insertion at position 210 and the upstream insertions at positions 211, 212, and 213 lie outside the pyrimidine gene cluster. This was verified by subsequent sequencing from the insertion at position 209 (see below). Colonies of E. coli carrying pBEM207 and pBEM208, with TnphoA insertions in pyrP (see below), were blue on LB agar containing XP, suggesting that the transposon insertions at positions 207 and 208 were in a gene encoding a protein exported in E. coli. Among the mgd insertion derivatives, pBEM218 and pBEM 219 did not complement pyrC, pyrAa, or pyrAb, while pBEM 220 and pBEM221 did not complement pyrD. pBEM217 did not complement the pyrE mutant, and pBEM222 did not complement the pyrAa mutant, the only strain with which it was tested. Since all insertion mutants complemented the E. coli pyrB mutant, there was a possibility that pKV48 contained a pyrB function outside the pyrimidine gene cluster. To test this, plasmid pBEM215 was used. This plasmid contains the left side of the insert in pKV48 and lacks most of the pyrimidine gene cluster, including the region found to encode pyrB (see below). Plasmid pBEM215 did not complement the E. coli pyrB mutant


VOL. 177, 1995

6869

TABLE 2. Complementation of E. coli auxotrophs by pKV48 and its transposon insertion derivatives Complementation of E. coli mutant carryinga: Plasmid

KV48 pBEM210 pBEM209 pBEM218 pBEM207 pBEM208 pBEM219 pBEM222 pBEM206 pBEM205 pBEM204 pBEM203 pBEM220 pBEM221 pBEM202 pBEM201 pBEM217

Insertion

None pKV48::TnphoA210 pKV48::TnphoA209b pyrR::mgd218 pyrP::TnphoA207 pyrP::TnphoA208 pyrC::mgd219 pyrAa::mgd222 pyrAb::TnphoA206 pyrAb::TnphoA205 orf2::TnphoA204 pyrD::TnphoA203 pyrD::mgd220 pyrD::mgd221 pyrD::TnphoA202 pyrF::TnphoA201 pyrE::mgd217

pyrB

pyrC

pyrAa

pyrAb

pyrD

pyrF

pyrE

1 NT 1 1 1 1 1 NT 1 1 NT 1 1 1 1 1 1

1 1 2 2 2 2 2 NT 1 1 NT 1 1 1 1 1 1

1 NT 2 2 2 2 2 2 2 2 NT 1 1 1 1 1 1

1 NT 2 2 2 2 2 NT 2 2 NT 1 1 1 1 1 NT

1 1 1 1 1 1 1 NT 1 1 1 2 2 2 2 1 1

1 NT 1 1 1 1 1 NT 1 1 NT 1 1 1 1 2 1

1 NT 1 1 1 1 1 NT 1 1 NT 1 1 1 1 1 2

a Complementation was performed by introducing pKV48 or its transposon insertion derivatives into E. coli auxotrophs and scoring for the ability to grow on M63 minimal agar. 1, growth observed; 2, no growth; NT, not tested. pyrAa and pyrAb are also called carA and carB, respectively, in E. coli. b This insertion is about 40 nucleotides upstream from the pyrP gene.

(data not shown), making it unlikely that a pyrB gene is encoded outside of the pyrimidine gene cluster on pKV48. Restriction mapping and sequencing of pKV48. The restriction map of pKV48 was generated by using EcoRI, BamHI, ClaI, or HindIII digestion of pKV48 and the 13 TnphoA insertion derivatives (Fig. 1). In most cases, both single and double digestions, and occasionally triple digestions, were performed. To facilitate sequencing and ordering of the genes contained in pKV48, three subclones, pBEM214, pBEM215, and pBEM 216, were constructed (Fig. 1). Subclone pBEM214 contains a 3-kb EcoRI-BamHI fragment of pKV48 spanning the TnphoA insertions at positions 202 and 203 which had resulted in loss of pyrD complementation. Subclone pBEM215 was generated by deleting two adjacent BamHI fragments (11 and 4.7 kb) from pBEM207, which is pKV48 with a TnphoA insertion at position 207. The 11-kb deleted BamHI fragment includes one arm of the inserted TnphoA. The remaining large fragment (about 37 kb) containing the other arm of TnphoA, including its kanamycin resistance gene, was self-ligated. An additional clone was constructed by excising and purifying the 11-kb BamHI fragment from pBEM207 and further digesting it with NotI and HindIII. A 2.5-kb NotI-HindIII fragment was ligated into appropriately digested pBluescript to produce pBEM216. Both pBEM215 and pBEM216 contain one of the TnphoA207 inverted repeats and thus contain sequences complementary to the GW26 primer, which was used to prime DNA sequencing reactions. Enterococcal sequences contained in pBEM216 and pBEM215 are contiguous to each other at position 207. The subclone pBEM214 was first sequenced from one end, by using the pBluescript T3 primer. The first 225 nucleotides demonstrated homology with the orf2 (open reading frame 2) sequence from Bacillus subtilis and B. caldolyticus, whose pyr genes are organized in similar clusters (14, 33). New primers were designed on the basis of the DNA sequence obtained and were used to extend the sequencing. In this way, both strands of the entire pyrD gene as well as part of the pyrF gene were sequenced. Most of the pyrD sequence was also obtained by manual sequencing of both strands. The pyrD gene is 936 nucleotides long in B. subtilis, while the enterococcal pyrD sequence is either 939 or 936 nucleotides long, depending on

which of two adjacent ATG codons is used for initiation of translation. The predicted E. faecalis PyrD amino acid sequence showed 79 to 81% similarity to PyrD from B. subtilis and B. caldolyticus and 53% similarity to PyrD from E. coli, reflecting the closer phylogenetic relationship among grampositive organisms. Recently, another gram-positive organism, Lactococcus lactis, was shown to contain two pyrD genes (1). The E. faecalis PyrD protein shows 88% similarity (73% identity) to PyrDb of L. lactis. The initial 270-nucleotide sequence from the subclone pBEM216, obtained by using the GW26 primer, revealed homology with pyrP of B. subtilis and B. caldolyticus (13, 37). Using a primer based on the pyrP sequence, we found that the next 300-nucleotide sequence had homology with the pyrB gene of these organisms. Sequencing from the other end of the 2.5-kb insert in pBEM216 by using the pBluescript T7 primer revealed a region of sequence similarity to pyrC, located about 5.5 kb upstream from pyrD in pKV48. Sequencing of about 200 nucleotides from pBEM215 by using the GW26 primer verified the expected pyrP sequence, which was continuous with the downstream pyrP sequence in pBEM216. The pyrP sequence was also obtained from direct sequencing of pBEM208 as described above. Using the phoA primer to sequence the region flanking TnphoA in cosmid pBEM206, which had lost the ability to complement pyrAb, we obtained 254 nucleotides of sequence similar to pyrAb of B. subtilis and B. caldolyticus. Similarly, sequence from the mgd insertion in cosmid pBEM222, which had lost the ability to complement pyrAa, provided a 310-nucleotide sequence that was similar to pyrAa of B. subtilis and B. caldolyticus. The pyrR homolog was sequenced from the insertion at position 209 by using the phoA primer and from the downstream insertion at position 218 by using the mgd-R primer. Computer analysis showed that the insertion at position 209 is about 40 nucleotides upstream of the predicted pyrR gene. Sequences flanking mgd insertions at positions 217 and 219 revealed similarity to pyrE and pyrC, respectively, of Bacillus spp. (Fig. 1). Insertions at positions 220 and 221 were about 120 nucleotides apart. In all, the lengths of the 10 sequenced regions (pyrR, pyrP, pyrB, pyrC, pyrAa, pyrAb, orf2, pyrD, pyrF, and pyrE) varied from 180 (pyrF) to 939 (pyrD)

6870

LI ET AL.

J. BACTERIOL.

FIG. 1. Analysis of pKV48. The large rectangle indicates 22 kb of enterococcal DNA, which contains the pyrimidine gene cluster. The pyr genes are indicated in the rectangle (R 5 pyrR, P 5 pyrP, etc.) by analogy with the pyrimidine gene cluster of B. subtilis. The pyrimidine pathway and enzymes encoded by these genes are shown in Fig. 3. Subclones are indicated by the smaller rectangles. pBEM215 was constructed by deleting two adjacent BamHI fragments from pBEM207. A 2.5-kb NotI-HindIII fragment from pBEM207 was subcloned into pBluescript to generate pBEM216. pBEM214 contained a 3-kb BamHI-EcoRI fragment from pKV48. The thick lines at the ends of pKV48 and pBEM215 represent cosmid pLAFRx. The thin lines flanking pBEM214 and pBEM215 represent pBluescript. The shaded bars within the pKV48 rectangle indicate regions that have been sequenced. Circles labeled 201 to 213 refer to TnphoA insertions, with relative orientations represented by open and filled circles. The structure of the TnphoA transposon is shown at the bottom. The orientation of the transposon shown is that corresponding to the filled circles. Triangles labeled 217 to 222 represent positions of mgd insertions, with relative orientations represented by open and filled triangles. The structure of the mgd transposon is shown at the bottom. The orientation of the transposon shown is that corresponding to the filled triangles. The location of the mgd-R primer is shown. Restriction enzyme cleavage sites: RI, EcoRI; B, BamHI; H, HindIII; CI, ClaI; Ev, EcoRV; X, XbaI.

nucleotides, and their DNA sequence similarity to the corresponding genes of B. subtilis and B. caldolyticus varied from 50 to 69%. The partial pyrP nucleotide and predicted peptide sequences were compared with the E. coli uracil permease sequence; the 296-nucleotide sequences showed 53% identity, and the predicted peptide sequences showed 78.3% similarity (data not shown). According to the map and sequence information, the distance from pyrB to pyrD in pKV48 is approximately 7.5 kb, while that between pyrB and pyrD in B. subtilis and B. caldolyticus is 7.2 kb. The distance between pyrC and pyrD and the distance between pyrAb and pyrD in E. faecalis were also very close to those in B. subtilis and B. caldolyticus. The alignment of the pyr gene cluster of these three organisms demonstrated that the organization in E. faecalis is very similar to that in B. subtilis and B. caldolyticus. Using the complete sequences from Bacillus spp., we determined the possible positions of the insertions in the pyr genes (Table 3). Isolation and characterization of insertions in the purL gene. The cosmid pKV53 was previously shown to complement an E. coli purL auxotroph (27). Using mgd, we performed mutagenesis with this cosmid. When the pool of pKV53 with mgd insertions was transduced into an E. coli purL auxotroph and selected as before, a colony with a mutant cosmid, designated pTX5062, that no longer complemented the E. coli auxotroph was isolated. Insertion of the transposon into purL of pTX5062 was confirmed by sequence analysis using the mgd-R

primer to prime the sequencing reaction. The purL gene of B. subtilis encodes a 228-amino-acid-long polypeptide. The transposon insertion in the E. faecalis purL gene was at a position corresponding to codon 21 (Table 3). Allelic replacement. We next tested the possibility of gener-

TABLE 3. Locations of transposon insertions in pyr and pur genes and growth of OG1 mutants Gene

Transposon insertion

Protein sizea (amino acids)

Location of insertiona,b (codon)

Phenotype of insertion in OG1RF

pyrR pyrP pyrC pyrAa pyrAb pyrD pyrD pyrE purL pyr cluster

mgd218 TnphoA208 mgd219 mgd222 TnphoA206 mgd220 mgd221 mgd217 mgd5062 mgdC22

182 435 429 367 1,072 313 313 217 228 NA

134 390 288 213 300 197 238 27 21 NA

Pyr1/2 NA Pyr2 NA NA Pyr1 Pyr1 Pyr2 Pur2 Pyr2

a

Based on the B. subtilis pyrimidine biosynthesis operon. Predicted by comparing sequences with pyr genes from B. subtilis. Pyr2 or Pur2, pyr or pur auxotrophic phenotype (optical density at 12 h, 0.03 to 0.07); Pyr1, prototrophic phenotype (optical density at 12 h, 0.21 to 0.26); Pyr1/2, slow growth in defined medium (optical density at 12 h, 0.12); NA, not available. b

VOL. 177, 1995

FIG. 2. Hybridization of genomic DNA from enterococcal transposon insertion mutants. A Southern blot of BamHI-EcoRI-digested DNA from TX5066b (lane 1), TX5066a (lane 2), OG1RF (lane 3), pKV48 (lane 4) and pBEM220 (lane 5) was hybridized with a 1-kb pyrD PCR product generated with primers just outside pyrD. TX5066b is a transposon insertion mutant generated by electroporating linearized pBEM220 DNA into OG1RF.

ating pyr or pur auxotrophs of enterococci by using the insertion mutants described above. Cosmids containing mgd insertions in the pyrR, pyrC, pyrD (two different insertions), pyrE, and purL genes (Table 2) were electroporated into OG1RF competent cells. Transformants that survived on kanamycin (2,000 mg/ml) were expected to result from recombination between the cosmid and the host chromosome because pLAFRx cannot replicate in enterococci. Southern blot analysis confirmed the locations of the transposon insertions in the chromosome of E. faecalis and showed that only a single insertion was present. Transformants were tested for auxotrophy by measuring growth on DMMS with or without uracil or adenosine supplements (Table 3). The transposon insertion mutants from pyrC, pyrE, and purL mutants grew poorly on DMMS agar, forming microcolonies, as well as in DMMS broth unless uracil or adenosine was added. The pyrR mutant showed intermediate growth under these conditions, possibly because of an incomplete polar effect on downstream genes. In contrast, both pyrD mutants showed growth that was comparable to that of the wild-type parent on DMMS agar or in broth. The insertion mutations in E. faecalis were stable, as no revertants were observed for pyrE or purL (frequency of ,1028) when these mutants were plated on DMMS agar. A single revertant of the pyrC insertion (TX5078), which grew well on DMMS agar, was detected at a frequency of less than 1028. The revertant was still resistant to 2,000 mg of kanamycin per ml, suggesting that the mgd transposon was still present. Presumably a rare second-site mutation was responsible for the phenotype. Because the pyrD insertions did not produce auxotrophy in enterococci, the mgd220 insertion mutant was analyzed in more detail. Thirteen putative OG1RFpyrD::mgd220 isolates hybridized with a probe from mgd but not with pLAFRx, indicating that kanamycin-resistant cells arose from a double crossover between the cosmid and the host chromosome (data not shown). Southern blots of DNA from two pyrD::mgd220 mutants (strains TX5066a and TX5066b) digested with BamHI plus EcoRI were hybridized with the 1-kb pyrD probe, and the results confirmed the correct allele replacement (Fig. 1 and 2). Finally, PCR of TX5066a was performed with a primer derived from the end of pyrD and a primer from within mgd. The amplified fragment was subjected to DNA sequence analysis, which verified that the mgd insertion was located in the pyrD gene. These results allowed us to conclude that insertion of mgd in the pyrD gene does not create an auxotrophic phenotype in E. faecalis. Chromosome replacement en masse. We wished to determine if a more efficient method for identification of biosynthetic genes (or any phenotypic class) could be performed in


6871

enterococci. To this end, we pooled several hundred mgd insertions in pKV48, isolated DNA, and electroporated the DNA into E. faecalis OG1RF. Twenty-three kanamycin-resistant colonies were analyzed, and one was found to have a pyrimidine requirement (Table 3). Southern blot analysis of this mutant (TX5081) showed it to have an insertion in the pyr gene cluster (data not shown), although the specific gene location was not determined. Thus, one can identify specific E. faecalis functions (e.g., metabolic genes) by this procedure without relying on heterologous complementation or other laborious procedures. DISCUSSION We show in this report that following transposon mutagenesis in E. coli, it is possible to construct auxotrophic knockout mutants of E. faecalis. Both pyrimidine- and purine-requiring auxotrophs, the first auxotrophic mutants of an enterococcal strain, were generated by this procedure. Clearly, mutations in any gene giving a phenotype that is detectable in E. coli could be isolated by this procedure. This should have particular applicability to genes involved in infection or other host interactions. Such genes could be detected in E. coli by using antisera from patients or other procedures. The genes coding for de novo pyrimidine biosynthesis have been extensively studied in E. coli and B. subtilis. In E. coli, the pyr genes are scattered around the chromosome and are not coordinately regulated (28). Pyrimidine-mediated regulation of expression of pyrBI and pyrE in E. coli has been shown to occur primarily by transcriptional attenuation (20). In B. subtilis, the genes encoding enzymes in the pyrimidine biosynthesis pathway are clustered on the chromosome and appear to be coordinately regulated (29, 31, 37). Expression is believed to occur from a promoter upstream from pyrR, producing a single transcript, that is negatively regulated by pyrR (37). At least one of these genes, pyrB, which encodes aspartate transcarbamylase, is regulated developmentally (23). Little was previously known about the organization of pyr genes or their regulation in enterococci. In this study, we used a defined medium, developed previously in this laboratory (27), to identify auxotrophs. It is clear that E. faecalis resembles Bacillus spp. with respect to organization of the pyr gene cluster. This resemblance is supported by the functional analysis of several genes and by complementation of E. coli mutants. It should also be noted that two TnphoA insertions into the pyrP gene, thought to encode a permease, gave blue colonies on XP medium, indicating export of the enterococcal pyrP gene product in E. coli. Thus, it is likely that this function is also conserved. The regulation of expression of the pyr gene cluster in E. faecalis also appears to share features with the same regulation in Bacillus spp. Thus, an insertion in pyrR (mgd218) showed reduced growth in DMMS, consistent with a polar effect of this insertion on downstream pyr genes. If pyrR is a negative regulator as in Bacillus spp., we would expect overexpression in addition to polar effects on downstream genes. Perhaps this explains the partial auxotrophic phenotype. Results of complementation in E. coli are more difficult to interpret, presumably because of the heterologous nature of the system. Insertions in pyrR, pyrP, or pyrC are deficient in complementation for downstream functions (except for pyrB) down to the pyrAa and pyrAb genes (Table 2). This finding is consistent with expression in E. coli of these enterococcal genes from a single promoter, upstream from pyrR. Note that complementation by these genes is also defective in the mutant with TnphoA209, which is inserted upstream from pyrR, con-

6872

LI ET AL.

J. BACTERIOL. ACKNOWLEDGMENTS This work was supported by NIH grant R01 AI33516 and by a grant from Pfizer, Inc. We thank David Needleman for assistance with automated sequencing.

FIG. 3. De novo synthesis of UMP. Genes referred to in Fig. 1 are shown and encode the following enzymes: pyrAa and pyrAb, glutaminase of carbamoylphosphate synthetase and carbamoylphosphate synthetase, respectively (also known as carAB [carbamoylphosphate synthase] in E. coli); pyrB, aspartate transcarbamylase; pyrC, dihydroorotase; pyrD, dihydroorotate dehydrogenase; pyrE, OMP-PRPP transferase; and pyrF, OMP decarboxylase.

sistent with the upstream location of the promoter. The pyrB gene in pKV48 derivatives containing upstream transposon insertions, however, complements an E. coli mutant despite the fact that its expression should be reduced in some of these mutants. The reason for this anomaly is not known. Another unusual result is the loss of complementation for pyrAa in an insertion mutant in the pyrAb gene (TnphoA206; Table 2). Effects of pyrAb knockout mutations on pyrAa have been reported previously (15) and may reflect the subunit nature of this enzyme. Another anomaly in the heterologous system is the complementation of the E. coli pyrD, pyrF, and pyrE mutants by these mutant cosmids (Table 2). Similarly, E. coli pyrF and pyrE mutants are complemented by cosmids with insertions in the pyrD gene (Table 2). Assuming that mgd and TnphoA insertions show polarity, these results are not consistent with a single transcript being produced in E. coli. Perhaps there are internal promoters in the pyr gene cluster that function in E. coli. Whether such promoters are physiologically significant in E. faecalis is not known. It is also possible that the anomalous complementation of pyrB, pyrD, pyrF, and pyrE reflects a requirement for smaller amounts of these gene products than for the other pyr functions. The observation that two independent pyrD knockout mutants, located centrally in the coding sequence, were not auxotrophs suggests that the PyrD function in E. faecalis may be duplicated, as in L. lactis. The two proteins in L. lactis show only 24% sequence identity, with PyrDa being related to the enzyme from Saccharomyces cerevisiae (70% identity) and PyrDb being related to the enzyme from B. subtilis (64% identity) (1). If this situation held for E. faecalis, it would explain why we did not detect the second gene by either Southern blot or PCR analysis, since the sequences of the E. faecalis pyrD and the L. lactis pyrDa genes are significantly different. Inspection of the pathway for de novo UMP synthesis (Fig. 3) shows that auxotrophs were isolated for steps before and after the reaction performed by the pyrD-encoded enzyme. The observation that auxotrophs were obtained for steps preceding the pyrDdependent reaction suggests that there is no alternative pathway for synthesizing orotic acid. Auxotrophs of E. faecalis may also be useful in studies of virulence. For example, in other systems, auxotrophs have been used to develop attenuated live vaccine strains (19). In addition, auxotrophic mutants were used to identify Salmonella genes that were expressed during infection (21). Such an in vivo expression procedure could be used for selection of enterococcal virulence genes that are specifically induced in host tissues.

REFERENCES 1. Andersen, P. S., P. J. G. Jansen, and K. Hammer. 1994. Two different dihydroorotate dehydrogenases in Lactococcus lactis. J. Bacteriol. 176:3975– 3982. 2. Bachmann, B. J. 1987. Derivations and genotypes of some mutant derivatives of Escherichia coli K-12, p. 1190–1219. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology. American Society for Microbiology, Washington, D.C. 3. Betzl, D., W. Ludwig, and K. H. Schleifer. 1990. Identification of lactococci and enterococci by colony hybridization with 23S rRNA-targeted oligonucleotide probes. Appl. Environ. Microbiol. 56:2927–2929. 4. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7:1513–1523. 5. Calvin, N. M., and P. C. Hanawalt. 1988. High-efficiency transformation of bacterial cells by electroporation. J. Bacteriol. 170:2796–2801. 6. Chung, C. T., S. L. Niemela, and R. H. Miller. 1989. One-step preparation of competent Escherichia coli: transformation and storage of bacterial cells in the same solution. Proc. Natl. Acad. Sci. USA 86:2172–2175. 7. Cruz-Rodz, A. L., and M. S. Gilmore. 1990. High efficiency introduction of plasmid DNA into glycine treated Enterococcus faecalis by electroporation. Mol. Gen. Genet. 224:152–154. 8. Dunny, G., C. Funk, and J. Adsit. 1981. Direct stimulation of the transfer of antibiotic resistance by sex pheromones in Streptococcus faecalis. Plasmid 6: 270–278. 9. Dunny, G. M., L. N. Lee, and D. J. LeBlanc. 1991. Improved electroporation and cloning vector system for gram-positive bacteria. Appl. Environ. Microbiol. 57:1194–1201. 10. Figurski, D. H., and E. R. Helinski. 1979. Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function provided in trans. Proc. Natl. Acad. Sci. USA 76:1648–1652. 11. Fogg, G. C., C. M. Gibson, and M. G. Caparon. 1994. The identification of rofA, a positive-acting regulatory component of prtF expression: use of an mgd-based shuttle mutagenesis strategy in Streptococcus pyogenes. Mol. Microbiol. 11:671–684. 12. Friedman, A. M., S. R. Long, S. E. Brown, W. J. Buikema, and F. Ausubel. 1982. Construction of a broad host range cosmid cloning vector and its use in the genetic analysis of Rhizobium mutants. Gene 18:289–296. 13. Ghim, S. Y., and J. Neuhard. 1994. The pyrimidine biosynthesis operon of the thermophile Bacillus caldolyticus includes genes for uracil phosphoribosyltransferase and uracil permease. J. Bacteriol. 176:3698–3707. 14. Ghim, S. Y., P. Nielsen, and J. Neuhard. 1994. Molecular characterization of pyrimidine biosynthesis genes from the thermophile Bacillus caldolyticus. Microbiology 140:479–491. 15. Gigot, D., M. Crabeel, A. Feller, D. Charlier, W. Lissens, N. Glansdorff, and A. Pie´rard. 1980. Patterns of polarity in the Escherichia coli carAB gene cluster. J. Bacteriol. 143:914–920. 16. Guyer, M. S., R. R. Reed, J. A. Steitz, and K. B. Low. 1981. Identification of a sex-factor-affinity site in E. coli as gamma delta. Cold Spring Harbor Symp. Quant. Biol. 45:135–140. 17. Heath, J. D., J. D. Perkins, B. Sharma, and G. M. Weinstock. 1992. NotI genomic cleavage map of Escherichia coli K-12 strain MG1655. J. Bacteriol. 174:558–567. 18. Highlander, S. K., M. Chidambaram, M. J. Engler, and G. M. Weinstock. 1989. DNA sequence of the Pasteurella haemolytica leukotoxin gene cluster. DNA Cell Biol. 8:15–28. 19. Kwaga, J. K., B. J. Allan, J. V. van der Hurk, H. Seida, and A. A. Potter. 1994. A carAB mutant of avian pathogenic Escherichia coli serogroup O2 is attenuated and effective as a live oral vaccine against colibacillosis in turkeys. Infect. Immun. 62:3766–3772. 20. Liu, C. G., and C. L. Turnbough, Jr. 1989. Multiple control mechanisms for pyrimidine-mediated regulation of pyrBI operon expression in Escherichia coli K-12. J. Bacteriol. 171:3337–3342. 21. Mahan, M. J., J. M. Slauch, and J. J. Mekalanos. 1993. Selection of bacterial virulence genes that are specifically induced in host tissues. Science 259:686– 688. 22. Manoil, C., and J. Beckwith. 1985. TnphoA: a transposon probe for protein export signals. Proc. Natl. Acad. Sci. USA 82:8129–8133. 23. Maurizi, M. R., and R. L. Switzer. 1978. Aspartate transcarbamylase synthesis ceases prior to inactivation of the enzyme in Bacillus subtilis. J. Bacteriol. 135:943–951. 24. Mergeay, M., D. Gigot, J. Beckmann, N. Glansdorff, and A. Pierard. 1974. Physiology and genetics of carbamoylphosphate synthesis in Escherichia coli K-12. Mol. Gen. Genet. 133:299–316. 25. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

VOL. 177, 1995 26. Murray, B. E. 1990. The life and times of the enterococcus. Clin. Microbiol. 3:46–65. 27. Murray, B. E., K. V. Singh, R. P. Ross, J. D. Heath, G. M. Dunny, and G. M. Weinstock. 1993. Generation of restriction map of Enterococcus faecalis OG1 and investigation of growth requirements and regions encoding biosynthetic function. J. Bacteriol. 175:5216–5223. 28. O’Donovan, G. A., and J. Neuhard. 1970. Pyrimidine metabolism in microorganisms. Bacteriol. Rev. 34:278–343. 29. Paulus, T. J., T. J. McGarry, P. G. Shekelle, S. Rosenzweig, and R. L. Switzer. 1982. Coordinate synthesis of the enzymes of pyrimidine biosynthesis in Bacillus subtilis. J. Bacteriol. 149:775–778. 30. Pearson, W. R., and D. J. Lipman. 1988. Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. USA 85:2444–2448. 31. Potvin, B. W., R. J. J. Kelleher, and H. Gooder. 1975. Pyrimidine biosynthetic pathway of Bacillus subtilis. J. Bacteriol. 123:604–615. 32. Prober, J. M., G. L. Trainor, R. J. Dam, F. W. Hobbs, C. W. Robertson, R. J. Zagursky, A. J. Cocuzza, M. A. Jensen, and K. Baumeister. 1987. A system for rapid DNA sequencing with fluorescent chain-terminating dideoxynucle-


6873

otides. Science 238:336–341. 33. Quinn, C. L., B. T. Stephenson, and R. L. Switzer. 1991. Functional organization and nucleotide sequence of the Bacillus subtilis pyrimidine biosynthetic operon. J. Biol. Chem. 266:9113–9127. 34. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 35. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463–5467. 36. Short, J. M., J. M. Fernandez, J. A. Sorge, and W. D. Huse. 1988. LambdaZAP: a bacteriophage lambda expression vector with in vivo excision properties. Nucleic Acids Res. 16:7583–7600. 37. Turner, R. J., Y. Lu, and R. L. Switzer. 1994. Regulation of the Bacillus subtilis pyrimidine biosynthetic (pyr) gene cluster by an autogenous transcriptional attenuation mechanism. J. Bacteriol. 176:3708–3722. 38. Zassenhaus, H. P., R. A. Butow, and Y. P. Hannon. 1982. Rapid electroelution of nucleic acids from agarose and acrylamide gels. Anal. Biochem. 125:125–130.